MME-MDD : um método para manutenção e evolução de sistemas baseados no MDD

• Ano de defesa: 2016
• Autor(a) principal: Arndt, Bruno Felipe
• Orientador(a): Lucrédio, Daniel
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC
• Palavras-chave em Português: Gerador de código; Migração de código; Método orientado a geração de código; Sincronização
• Palavras-chave em Inglês: Code generator; Code migration; Code generation method; Synchronization; MDD (Model-Driven Development)
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::TEORIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/8503

Model-Driven Development (MDD) is proposed to reduce the semantic gap between problem and solution/implementation domains. Some tools are used, and the code generator is commonly used in this context. These generators are often implemented using templates. A Reference Implementation (RI) favoring the development/ maintenance of software facilitates this implementation’s types. However, RI requires a code migration process which consists in artifacts’ synchronization, and it is responsible for 20% to 25% of the time spent on development. The literature has no describe automatic solutions, but the group that this reaserch was included has develop tools that automate this process and reduce the time spent on some tasks. But each task has a different performance in relation to time spent, so automation of some tasks can be disadvantageous. Few reports describe such tasks and which ones are capable of automatic code migration. The aim of this study is to investigate the maintain and evolution process to identify and describe the types of maintain and evolution tasks that are essentially. Based on this study, a method (MME-MDD) that drives the developer during that proccess to guide the developer during the realization of each task, with the aim to maximize the benefits of this approach. The MME-MDD was validated by a case study and a empirical study and the method showed effective in most of tasks. In addition, studies show that using the proposed method brought a gain in the quality of the source code.